The idea that could triple radar range: DARPA’s THREADS programme moves into Phase 2

A radar’s real limit is thermal, not geometric

How far a radar sees is not, as often assumed, purely a function of aperture size. At the other end of the equation sits transmitted power — and the moment you raise power you hit a wall made of heat. Every watt radiated comes with a share converted to heat inside the chip, and if that heat cannot leave, the device first loses performance and then loses life.

This is why the hardest problem in a modern active electronically scanned array is not signal processing but cooling. Liquid loops, heat pipes, cold plates — all of them try to catch the heat after it has already left the die. DARPA’s THREADS programme asks a different question: can the heat be taken away before it spreads, micrometres from the transistor that made it?

The acronym says exactly that — Technologies for Heat Removal in Electronics at the Device Scale. The phrase “device scale” is the operative one: not the enclosure, not the board, but the semiconductor layer itself.

Why gallium nitride

Gallium nitride has been the quiet revolution in defence electronics for fifteen years. Compared with gallium arsenide or silicon-based amplifiers it operates at far higher voltage and power density, which means far more transmitted power from the same volume. The world’s leading AESA radars, satellite communications terminals and electronic warfare jammers are built on it.

But GaN’s advantage is also its constraint. Push more power through the same area and the heat flux climbs with it. It is an open secret in the field that GaN devices are routinely operated well below their theoretical capability, because at full drive the heat generated cannot be safely carried away. THREADS is aimed squarely at that gap: not inventing a new semiconductor, but running the existing one all the way to its design limit.

Isaac Wildeson, principal investigator at BAE Systems’ FAST Labs, summarised the first phase this way: “The progress we’ve made during Phase 1 validates our approach to material and process enhancements and brings us closer to unlocking the full potential of RF-based systems for our warfighters.”

DARPA THREADS Programme Profile

What tripling the range would mean

“Three times the range” is a claim of a size rarely heard in defence technology. Made concrete: a radar detecting at 200 kilometres today would reach 600, or could hold the same range from a far smaller, lower-power aperture. In electronic warfare it means a jamming aircraft doing its job without ever entering the engagement envelope of the air defence system it is working against.

The attractive part is where the gain comes from. No new platform, no new array, no new waveform — it is already latent in the chip being manufactured today. In defence procurement, running the same box harder is usually the cheapest capability increase available, and the fact that it applies directly to modernising existing inventory is what gives programmes like this a privileged seat at the budget table.

The caveat belongs here too: THREADS is still a research programme. Phase 2 will test whether an approach validated in the laboratory becomes a manufacturable process. The history of semiconductors is not short of ideas that worked on the bench and stumbled in the fab.

Beyond the US: GaN as a sovereignty question

The story reads differently outside Washington, because GaN now sits at the base of almost every serious defence electronics industry. In Türkiye’s case it underpins a broad ASELSAN product family, from the air defence radars supporting the SİPER and HİSAR systems to naval multifunction radars and electronic warfare jammers. Ankara has been investing for years in moving GaN from a purchased component to a domestically produced one — a choice whose value compounds every time export controls make the news.

What programmes like THREADS demonstrate is that owning GaN is the start of the race rather than the finish. The next round of competition will be decided by who can safely draw the most power from the same material. Building a domestic GaN line buys the ticket; how many laps that ticket covers depends on what is spent on materials and thermal research.

Sources

• Defence Industry Europe — “BAE Systems advances DARPA THREADS effort to Phase 2”, 15 August 2026
• BAE Systems FAST Labs corporate statements
• DARPA Microsystems Technology Office programme descriptions
• Wikipedia — Gallium nitride, Active electronically scanned array

Related Coverage

• US Space Force and Japan complete geosynchronous surveillance payloads
• US Marines open first counter-drone live-fire lanes
• MQ-25 Stingray hit by four-year delay and a $19.4 billion bill